I'm a Business Analytics graduate student at The George Washington University, driven by a passion for machine learning , ethical AI, and solving complex real-world problems through data. With a multidisciplinary background in marketing, fitness entrepreneurship, and analytics, I combine technical skills with strategic insight to create impactful, data-informed solutions.

• Designing interpretable and bias-aware ML models using SHAP, EBM, and AIR
• Building scalable, real-time forecasting tools using Python, SQL, and cloud technologies
• Exploring CECL (Current Expected Credit Loss) frameworks to develop responsible credit risk models
• Bridging data science and business strategy through visualization and experimentation
• Applying the discipline and resilience I’ve developed through years of weightlifting and coaching to my work in analytics

Languages & Platforms
Python • R • SQL • PostgreSQL • AWS • Apache Spark

Visualization
Power BI • ggplot2 • SAS Visual Analytics

Analytics & Modeling

• Supervised ML: Regression, SVM, LASSO, Ridge
• Explainable ML: SHAP, EBM, Post Hoc Explanation
• Statistics: ANOVA, A/B Testing, Forecasting
• Financial Modeling: CECL components (PD, LGD, EAD)

• Capital Bikeshare Demand Forecasting
  Developed LASSO and Ridge models with 85% accuracy to optimize daily fleet planning.

• Financial & Risk Analytics: CECL, Credit Risk, and Predictive Modeling Built a CECL-compliant forecasting framework in R, combining ARIMAX time-series models with macroeconomic drivers: unemployment, home prices, and delinquency rates.

• Fair Lending Bias-Remediated ML
  Created transparent, bias-remediated models for fair mortgage decision-making using SHAP and EBM.

• Iowa Liquor Sales Market Entry Built a complete SQL + Python pipeline on AWS with Power BI dashboards to identify underserved regions.

LinkedIn https://www.linkedin.com/in/maryamshahbazali/

GitHub: https://github.com/tsjmaryam

Portfolio: https://tsjmaryam.github.io/